Rethinking the forecasting of innovation diffusion: A combined actor- and system approach

Technological forecasting has significantly expanded over the last decades, leading to widespread use of forecasting models for explaining technology adoption and diffusion of innovation. While these models are broadly used, they have faced criticism for narrowing the explanatory components of adoption, focusing on adopters, innovation characteristics, or environmental factors, but seldom combine these to address complex problems holistically. This paper aims to combine actor- and system perspectives on innovation diffusion with the intention to broaden the explanatory power of traditional forecasting models. The study focuses on the case of solar photovoltaic (PV) diffusion in Sweden, surveying 46,507 residential PV adopters that applied for the capital subsidy program between 2009 and 2021 about their adoption satisfaction. Findings suggest that traditional models primarily account for direct effects on adoption satisfaction, whereas incorporating system-level factors captures indirect effects, providing a more comprehensive understanding of technology adoption. This highlights the interplay between actor- and system-level factors and acknowledging the holistic nature of innovation diffusion, which can inform future forecasting practices. • Traditional forecasting models are lacking a system-perspective • We broaden traditional forecasting models combining actor- and system-perspectives • Traditional models primarily account for direct effects on technology adoption • Incorporating system-level factors captures indirect effects on technology adoption